Private Equity Bake-off: Buyout vs. Venture Capital

by Ben Lorica (last updated April/2012)

A recent paper by Harris, Jenkinson, and Kaplan (HJK)1 provided performance statistics for two major styles of private equity - buyout and venture capital funds. The authors computed results using data from six different databases. I've taken their summary tables and converted them into a series of simple charts, but if you're interested in private equity you definitely should read the entire paper.
1. Data providers: I limit myself to results based on data supplied by Burgiss ("sourced from over 200 institutional investors") and Venture Economics. The combination of Burgiss and Venture Economics seems to capture the essence of their results. HJK also computed results using data from Prequin, Cambridge Associates, and other databases.2

2. Summary Statistics: Mean and Median should be clear or easy to lookup. Weighted Average uses weights based on "the capital committed for each fund as a proportion of the total commitments, as estimated by each source, for each vintage year."

3. Metrics include market-adjusted performance (PME) and absolute performance (IRR and multiple of invested capital (TVPI)). All three metrics were computed net of fees: IRR stands for Internal Rate of Return, a topic I touched on in a separate article. TVPI stands for Total Value Paid-in and is also referred to as the Investment Multiple:

TVPI = Investment Multiple = (Cumulative Distributions + Residual Value) / ( Paid-In Capital )

TVPI = Investment Multiple = (All fund distributions + the value of unrealized investments) / ( All fund contributions by LPs )

PME stands for Public Market Equivalent: In the charts below PME compares "private equity returns to equivalent timed investments in the S&P 500". When PME is greater than one, the private equity investment beats the S&P 5002, net of fees.

4. Vintage year is the year a fund makes its first investment.

Before you play with the charts I've created, here are some themes that emerge from the summary tables found in the paper:

  • Regardless of the metric or data source used, Venture Capital underperformed Buyout funds over the most recent decade and in the 1980's. It's true the 1990's technology bubble led to stellar VC returns for late 1990's vintage years. Breaking out summary statistics for each decade highlights how exceptional the late 1990's truly were for Venture Capital funds.

  • Market-adjusted performance (PME's): Using the weighted-average of PME's, buyouts outperformed the S&P 500 in all but five vintage years (1985, 1992, 1996, 2006, 2008). In contrast Venture Capital underperformed the S&P 500 in 12 out 29 vintage years (1984-86, 1999-2004, 2006-2008). The contrast is even sharper4 when one looks at the median value of PME's: Buyouts underperformed the S&P 500 in 6 out of 29 vintage years, while Venture Capital underperformed the S&P 500 in 16 out of 29 vintage years.

  • TVPI should be preferred to IRR as a summary measure of private equity performance: Compared to IRR (internal rate of return), TVPI (or multiple of invested capital) is a more accurate indicator of market-adjusted performance. The authors built statistical models that showed that for a given vintage year, PME is reliably predicted by a fund's TVPI and IRR, with TVPI providing higher explanatory power. Either are much better than simple-minded, end-to-end returns.

    • [Drop-down menu lets you toggle between different combinations of metric and data source.]

    Metric - Data Source:

    Private Equity By Vintage Year: Buyout vs. Venture Capital
    [Source: Tables 1-6 from Private Equity Performance: What Do We Know?]

     
     
    Related resources:
  • Venture Capital Funds: Long-term Performance

  • When evaluating Investment Funds, use Dollar-weighted Returns

  • Hedge Fund profits flow mostly to Industry Insiders

  • Data Mining with the Maximal Information Coefficient

  • Hedge Fund Performance: My semi-regular blog posts on the Dow Jones / Credit Suisse Hedge Fund indices.


    • (1) Private Equity Performance: What Do We Know? by Robert S. Harris, Tim Jenkinson, and Steven N. Kaplan. The paper is discussed in this short article in the Economist. I also recommend an older expository article written by one of the researchers.

    (2) By comparing results across data providers, the researchers conclude that "it is highly likely that Venture Economics returns understate buyout fund performance".

    (3) The authors performed similar calculations using other public market benchmarks: Russell 3000, the Nasdaq, and the Russell 2000. Results were generally similar to ones for the S&P 500.

    (4) Using the MEDIAN PME, Buyouts underperformed the S&P 500 in 6 out of 29 vintage years: 1984-85, 1988, 1992, 2006, 2008. In comparison Venture Capital underperformed the S&P 500 in 16 out of 29 vintage years: 1984-86, 1989, 1998-2008!  
     
     
    Back to Resources page.


    NOTE: Reproduction & reuse allowed under Creative Commons Attribution.    Creative Commons Attribution